Real Time Computer Vision

Computer vision, one of the deep learning applications, was used in this work repository. Many studies have been carried out using OpenCV and mediapipe libraries and source codes have been added to the repository.

Files

Files	Description
Output_img	Sample output images
HandTracking.py	real-time detection of hand points
Hand_Module.py	Modulated script of HandTracking file
Volume_Hand_Control.py	volume control with finger spacing
face_mesh.py	detection of face points
pose_estimation.py	real-time detection of human poses
requirements.txt	required libraries
FingerCounting.py	Finger Counting Code
FingerImages	Finger Counting visualization

Results

1- Hand Tracking

With the help of the mediapipe library, the points on the hand are detected in real time. The output is a screenshot taken from the real-time image.

2- Hand volume control

With the help of the mediapipe library, the points on the hand are detected in real time. Then the sound level of the computer is adjusted with the distance between the fingers.

3- Face Mesh

The points on the face are detected in real time using mediapipe and opencv, and these points are combined and masked. The output is a screenshot from the real-time image.

4- Pose Estimation

real-time pose detection is performed by detecting points on the human body.